> For the complete documentation index, see [llms.txt](https://docs.opsmx.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.opsmx.com/remediation-agents/code-agent.md).

# Code Agent

AI Guardian is OpsMx's **AI-powered developer security tool** that finds, explains, and automatically fixes security vulnerabilities in your source code and dependencies — directly within your existing GitHub workflow. It connects to your repositories, runs SAST and SCA scans, and remediates issues by generating and raising pull requests automatically — all without requiring prior security expertise from the developer.

AI Guardian is the **pre-deployment security engine** in the OpsMx Dual-AI architecture — working alongside Argonaut (post-deployment operations) to create a complete, closed-loop security model from code to cloud.

## What Is AI Guardian

AI Guardian is a developer-facing security platform that:

* **Connects** to your GitHub repositories via a secure GitHub App integration.
* **Scans** source code and third-party dependencies automatically for known vulnerabilities.
* **Explains** each finding in plain language — no security expertise required.
* **Remediates** issues through an AI-powered interactive chat — generating code patches and raising pull requests in GitHub.
* **Monitors continuously** — auto-scanning repositories at configurable intervals so new vulnerabilities are caught as code evolves.
* **Secures pull requests** — automatically scanning every PR before it is merged, preventing new vulnerabilities from entering the main branch.

## Who Should Use AI Guardian

| Persona                 | How They Use AI Guardian                                                                                                     |
| ----------------------- | ---------------------------------------------------------------------------------------------------------------------------- |
| **Software Engineers**  | Get fast, inline feedback on code vulnerabilities — fix issues in minutes with AI-generated patches                          |
| **DevOps Teams**        | Automate security checks in CI/CD pipelines — PR scans and auto scans run continuously without manual intervention           |
| **Security Teams**      | Monitor and remediate risks across all repositories from a single dashboard — without requiring developers to switch tools   |
| **Product Owners**      | Ensure secure delivery without slowing development — AI Guardian runs in the background, surfacing only what needs attention |
| **Platform Architects** | Integrate AI-powered security into platform workflows — GitHub App integration requires no custom pipeline scripting         |

## Key Terms

| Term                   | Description                                                                                                                                        |
| ---------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Hub**                | A central workspace where you group related GitHub projects to manage scanning and remediation — useful for managing multiple GitHub organizations |
| **GitHub Integration** | A secure connection between AI Guardian and your GitHub account that allows access to repositories via a GitHub App                                |
| **Project**            | A specific GitHub repository and branch that you want to scan and monitor                                                                          |
| **SAST**               | Static Application Security Testing — scans source code for security vulnerabilities without running the code                                      |
| **SCA**                | Software Composition Analysis — scans third-party libraries and dependencies for known CVEs and license risks                                      |
| **AI Remediation**     | AI-generated fix for a detected security issue — reviewed in an interactive chat and applied via an automated pull request                         |
| **PR Scan**            | Automatic security scan triggered on every pull request — reports only new vulnerabilities introduced by that specific PR                          |
| **Auto Scan**          | Scheduled, recurring scans at configurable intervals — ensures continuous security monitoring without manual intervention                          |
| **Chat History**       | Persistent remediation session history — allows developers to resume an in-progress remediation if interrupted or logged out                       |